Speq Implement Pr logo

Speq Implement Pr

Community
marconae
speq-implement-pr

Builds a planned feature on its feat/plan-name branch as a follow-up to /speq-plan-pr: implements, bumps the version, runs the real test suites, records the specs if green, comments the verification, and marks the PR ready. Run again to resume. Arg: plan name, PR number, or branch name.

Overview

Publishermarconae
Repositoryspeq-skill
Skill namespeq-implement-pr
Stars
54
Forks
10
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by marconae on GitHub. Read the source before you install it.

Installation

Install the Speq Implement Pr AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/marconae/speq-skill/tree/main/.claude/skills/speq-implement-pr
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/marconae/speq-skill.git /tmp/speq-skill
mkdir -p .claude/skills
cp -r /tmp/speq-skill/.claude/skills/speq-implement-pr .claude/skills/speq-implement-pr
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Speq Implement Pr in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Speq Implement Pr on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Speq Implement Pr is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Spec Implementer, PR flow (Orchestrator)

You are a thin orchestrator on top of speq-implement. Drive a plan on its existing feat/<plan-name> branch to a ready PR through the existing skills, run unchanged:

  • /speq-implement builds and verifies. /speq-record merges the specs. You run every git/gh action yourself, per /speq-git-operations.
  • Record only after the project's real test suites are green.
  • Stop on the first failed step and report it. Running the skill again resumes the work.

Required Skills (for the orchestrator)

Invoke before starting:

  • /speq-cli: spec discovery, to resolve plan names
  • /speq-writing-guardrails: prose style for artifacts and GitHub text
  • /speq-git-operations: the git/gh operation-to-command mapping, safety rules, and return formats. You run every operation directly

speq-implement and speq-record invoke their own required skills.

Workflow

0. Load Project Hook

Check for .speq/implement-pr-hook.md in the repo root.

  • Present: read it. Announce "Loaded project hook: .speq/implement-pr-hook.md". Its content is authoritative: it can add to, change, or override any part of this workflow. If the hook conflicts with this workflow, the hook wins.
  • Absent: continue without mention.

1. Resolve Target

$1 empty → ask the user which plan. Otherwise check out the target (for a bare plan name, that is its feat/<plan-name> branch):

Run — operation: checkout (per /speq-git-operations)
  target: <$1: a plan name (→ feat/<plan-name>), PR number, or branch name>

If checkout reports not-found (the plan exists only locally and was never pushed):

Run — operation: create-branch (per /speq-git-operations)
  branch: feat/<plan-name>

If specs/_recorded/*-<plan-name>/ exists and specs/_plans/<plan-name>/ does not, the plan is already recorded: skip to step 6.

2. Implement

If verification-report.md exists in the plan directory and every task in tasks.md is [x], skip this step.

Otherwise invoke /speq-implement <plan-name> and let it run to completion. If tasks.md exists, it resumes from the first incomplete task.

3. Bump Version

Skip this step if the version file already differs from the default branch.

Read the tech stack in specs/mission.md and bump the version in that stack's version file (for example Cargo.toml for Rust, package.json for Node, pyproject.toml for Python). Apply the conventional next version per Conventional Commits: feat → minor, a purely fix plan → patch, a breaking change → major. Do not take the version from the plan's spec deltas. Run a build to keep the lockfile in sync, redirecting its output to a log: mkdir -p target && <build-command> > target/speq-build.log 2>&1. Branch on the exit code. If a report quotes output, quote at most tail -n 30 of the log.

4. Commit Evidence

Commit and push now, so the evidence artifacts reach git history before /speq-record archives the plan out of tracked space, and so the branch survives workspace loss. Derive <type>(<scope>) per /speq-plan-pr's PR-title derivation:

Run — operation: commit (per /speq-git-operations)
  paths: implementation files, version bump, and the whole specs/_plans/<plan-name>/
         directory, not an itemized subset, except specs/_plans/<plan-name>/notes/
         (local working state, never committed)
  message: <type>(<scope>): implement <plan-name>

Run — operation: push (per /speq-git-operations)

5. Test and Record

  1. Run the suites. Run the project's real test suites (per specs/mission.md § Commands, typically an integration suite and an end-to-end suite), redirecting each suite's output to a log: mkdir -p target && <suite-command> > target/speq-<suite>.log 2>&1. Judge green or red by exit code. If any suite is red, report the failed suites with at most tail -n 30 of the relevant log, stop, and leave the plan unrecorded. After the user fixes the code, run this skill again.
  2. Record. Invoke /speq-record <plan-name>. If it asks its library-threshold split question, answer yes (split). Recording accepts the ADR candidates on the plan PR. Your run is the user's go-ahead, so do not stop for an ADR question. If it returns Recording failed: <reason> (for example a stale architecture delta base), stop and report it. After the user fixes the delta or the plan, run this skill again.
  3. Note the Archive: path /speq-record returns.

6. Ship

  1. Ship. One composite call. Step 4 already committed the implementation and evidence, so this commit covers only what /speq-record produced: the merge results and the archive's removal of the plan directory. Derive <type>(<scope>) and <slug> per /speq-plan-pr's PR-title derivation. Do not rewrite the PR body speq-plan-pr set. Use Implements <plan-name>. as the body, which applies only if no PR exists yet:

    Run — operation: ship-ready (per /speq-git-operations)
      paths: permanent-spec merges (specs/<domain>/...), specs/architecture.md,
             specs/_decision/ additions, deletion of specs/_plans/<plan-name>/
      message: <type>(<scope>): record <plan-name>
      title: <type>(<scope>): <slug>
      body: Implements <plan-name>.

    ship-ready reuses the draft PR speq-plan-pr opened (or opens one if the plan was only implemented locally) and marks it ready. A commit with nothing to commit is a no-op, so running this step again is safe.

  2. Comment. Post the verification summary. Build it from <archive-path>/verification-report.md, where <archive-path> is the Archive: path from step 5 or, when step 1 skipped ahead, the match of specs/_recorded/*-<plan-name>/. Compose it per /speq-writing-guardrails and /speq-implement's references/implementation-summary-template.md:

    Run — operation: comment-pr (per /speq-git-operations)
      body: <the composed verification summary>
  3. Report the finished PR and leave it for the user's review. Never merge it.

Spec Hierarchy (reference)

specs/
├── <domain>/<feature>/spec.md            # Permanent (after record)
├── architecture.md                       # Permanent, changed by the recorder applying the plan's architecture delta
├── _plans/<plan-name>/                   # Active until recorded: committed whole at step 4 except notes/
│   ├── architecture.md                   # Architecture delta, present only when the plan changes the architecture
│   ├── tasks.md                          # Created by speq-implement
│   ├── review-findings.md                # Created by code-reviewer
│   ├── review/review.md                  # Created by plan-reviewer
│   ├── notes/<group-letter>.md           # Rotation hand-off notes, created by implementer agents; local, never committed
│   └── verification-report.md            # Created by speq-implement
└── _recorded/NNN-<plan-name>/            # Archived by speq-record (gitignored by default)

Work Split (reference)

StepPerformed by
Target resolution, version bump, coordinationThis skill (pins Sonnet)
Task breakdown, coding, reviewspeq-implement
Spec merge and archivespeq-record
Branch, commit, push, PR create/updateThis skill, directly, per /speq-git-operations

Frequently asked questions

What does the Speq Implement Pr AI skill do?

Builds a planned feature on its feat/plan-name branch as a follow-up to /speq-plan-pr: implements, bumps the version, runs the real test suites, records the specs if green, comments the verification, and marks the PR ready. Run again to resume. Arg: plan name, PR number, or branch name.

Why use Speq Implement Pr on TypingMind?

Because you install it once and use it with any model. Speq Implement Pr is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Speq Implement Pr in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/marconae/speq-skill/tree/main/.claude/skills/speq-implement-pr. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Speq Implement Pr?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Speq Implement Pr?

As many as you like. As long as a model supports skills, you can use Speq Implement Pr with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Speq Implement Pr AI skill free?

Yes. It is published on GitHub by marconae under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇